Process for the purification of a carboxylic acid-containing composition

Abstract

A carboxylic acid-containing composition, which composition contains an aldehyde, is purified in a process, which process comprises introducing the carboxylic acid-containing composition and an aqueous electrolyte into an electrolytic cell comprising electrodes; electrochemically oxidizing the aldehyde in the electrolytic cell to obtain an electrochemically oxidized product composition comprising a carboxylic acid derived from the aldehyde; and, optionally, separating carboxylic acid from the electrochemically oxidized product composition.

Claims

1. A method for the purification of a carboxylic acid-containing composition, which composition further contains an aldehyde, comprising: introducing the carboxylic acid-containing composition and an aqueous electrolyte into an electrolytic cell comprising electrodes; and electrochemically oxidizing the aldehyde in the electrolytic cell to obtain an electrochemically oxidized product composition comprising a carboxylic acid derived from the aldehyde, wherein the carboxylic acid has the formula HOOCAr.sup.1COOH and the aldehyde has the formula OHCAr.sup.1COOH, wherein Ar.sup.1 represents an arylene or heteroarylene moiety.

2. The method according to claim 1, wherein Ar.sup.1 is selected from phenylene, furylene and pyridylene moieties.

3. The method according to claim 1, wherein the electrolytic cell is a divided cell.

4. The method according to claim 1, wherein at least one of the electrodes comprises a non-noble metal and/or an oxide and/or a hydroxide thereof.

5. The method according to claim 1, wherein carbon is used as cathode material in the electrolytic cell.

6. The method according to claim 1, wherein the aqueous electrolyte comprises an alkaline solution.

7. The method according to claim 6, wherein the alkaline solution comprises an alkaline compound selected from an alkali metal hydroxide, alkali metal carbonate, alkali metal bicarbonate, ammonia, ammonium carbonate, ammonium bicarbonate, a trialkylamine and combinations thereof.

8. The method according to claim 1, wherein the potential difference between anode and cathode in the electrolytic cell is at most 10 V.

9. The method according to claim 1, wherein the aldehyde is oxidized at a temperature in the range of 10 to 250 C. and at a pressure in the range of 0.5 to 20 bar.

10. The method according to claim 1, wherein the residence time of the carboxylic acid-containing composition in the electrolytic cell is in the range of 0.1 to 24 hours.

11. The method according to claim 1, which method is conducted in a continuous mode.

12. The method according to claim 1, wherein the carboxylic acid derived from the aldehyde is recovered by acidizing the electrochemically oxidized product composition and allowing carboxylic acid to precipitate.

13. The method according to claim 1, further comprising separating the carboxylic acid derived from the aldehyde from the electrochemically oxidized product composition.

Description

EXAMPLE 1

(1) In order to mimic a purification of a carboxylic acid-containing composition that is obtained by the oxidation of HMF a solution comprising FFCA and FDCA, each in a concentration of 50 mmoles per liter 0.5 M NaOH in water, was made. A divided electrolytic cell consisting of two compartments separated from each other by means of a porous glass frit, was used. The solution comprising FFCA and FDCA was placed in one compartment, i.e. the anode compartment, of the divided electrolytic cell. The anode compartment was further provided with an anode, i.e. a nickel plate. The other compartment, i.e. the cathode compartment, was provided with an aqueous solution of 0.5M NaOH and a cathode consisting of a nickel mesh. Both compartments were stirred. At room temperature, i.e. about 20 C., a current was applied on the electrodes. The current was 6.4 mA, corresponding with a current density of 0.8 mA/cm.sup.2. The voltage measured at the anode was 0.4-0.7 V versus reference Ag/AgCl electrode. The current was continued for 6.7 hours. At the anode the FFCA was oxidized to FDCA. At the cathode hydrogen evolved. After 6.7 hours the content of the solution in the anode compartment was analyzed. The conversion of FFCA was measured as molar percentage of aldehydes that have disappeared. Apart from FFCA and FDCA, no other compounds were detected in the solution of the anode compartment. That means that the result of a Cannizarro reaction that may have taken place in the anode compartment forming 5-hydroxymethyl-furan-2-carboxylic acid (HMFCA) and FDCA was offset by the further oxidation of any HMFCA that is formed, at the anode to FDCA. Thereby the yield of FDCA is optimized. The aldehyde conversion is shown in Table 1 below.

(2) To show the suitability of the present process for other aldehydes, another experiment was conducted with 50 mM/L furfural and 50 mM/L furoic acid in 0.5 M NaOH. The reaction with furfural and furoic acid lasted 6.7 hours. The conversion of the aldehyde is shown in Table 1.

(3) TABLE-US-00001 TABLE 1 Experiment No. Reagent Aldehyde conversion, % 1 FFCA + FDCA 90.0 2 Furfural + furoic acid 71.9

(4) The experiments show that the electrochemical oxidation of an aldehyde that is contained in a carboxylic acid-containing composition leads to conversion of the aldehyde into the corresponding acid in a major proportion without leading to undesired by-products.

EXAMPLE 2

(5) A series of experiments were carried out in substantially the same way as described for the experiments in Example 1. The electrolyte was 0.5M NaOH solution. The feedstock and the amount thereof (in mmoles per liter NaOH solution) have been shown in Table 2. Table 2 also shows the reaction temperature as well as residence time of the feedstock in the electrolytic cell. The electrodes both consisted of nickel mesh. The current applied amounted to 22.4 mA, corresponding with a current density of 0.8 mA/cm.sup.2 and a potential at the anode of 0.4-0.8 V versus a reference Ag/AgCl electrode. In Experiment No. 4 a feedstock was used that consisted of crude FDCA, obtained in the oxidation of methoxymethylfurfural with oxygen in acetic acid using a Co, Mn and Br-containing catalyst. The crude FDCA contained about 1% wt FFCA, based on the total crude FDCA, and minor amounts of color bodies. The Table also shows the conversion of the aldehyde.

(6) TABLE-US-00002 TABLE 2 Reaction Exp. temperature, Residence Aldehyde No. Feedstock C. time, hr conversion, % 3 5 mM FFCA + 45 mM 20 1.7 100.0 FDCA 4 50 mM crude FDCA 20 1.0 100.0

(7) The above results show the suitability of the present process in the purification of mixtures of FDCA and FFCA. Whereas the feedstock of experiment No. 4 shows a brown/yellow color, the product after electrochemical oxidation is almost colorless, indicating that major color bodies have been removed.

EXAMPLE 3

(8) The use of an undivided electrolytic cell is also shown in Experiment Nos. 5-8. A glass vessel used as an undivided electrolytic cell was loaded with a solution of feedstock as indicated in Table 3 having a concentration of the number of millimoles indicated per liter aqueous 0.5 M NaOH, an anode and a cathode. The material of the anode was a nickel mesh as indicated in Example 2, and the cathode was made of nickel mesh or carbon paper. A current of 22.4 mA was applied between the anode and cathode. The electrochemical oxidation was conducted at room temperature, i.e. 20 C., for a period as shown as the residence time in Table 3. The feedstock, cathode material and aldehyde conversion in the aqueous electrolyte after the residence time indicated are also shown in Table 3. The electrolyte was also varied by using 0.5 M NaOH in water or 0.5 M triethylamine (TEA) in water. The feedstock in experiment Nos. 6-8 was crude FDCA, including 1% wt FFCA, based on the total crude FDCA and minor amounts of color bodies.

(9) TABLE-US-00003 TABLE 3 Aldehyde Exp. Cathode Residence conversion, No. Feedstock material Electrolyte time, hr % 5 50 mM FFCA + Ni mesh NaOH 5.6 97.9 50 mM FDCA 6 150 mM crude Ni mesh NaOH 5.6 98.4 FDCA 7 50 mM crude Carbon NaOH 3.5 100.0 FDCA paper 8 50 mM crude Carbon TEA 5.1 99.1 FDCA paper
In addition to a virtually complete conversion of the aldehyde compound, the resulting product in experiment Nos. 5-8 also showed considerably less coloring, indicating that also color bodies were oxidized.

EXAMPLE 4

(10) To show that the present process can also be applied to aldehydes other than FFCA, two further experiments were conducted on a feedstock comprising benzaldehyde and benzoic acid in one experiment and on 4-carboxybenzaldehyde (4-CBA) and terephthalic acid in the second experiment. The experiments were conducted in a way similar to the experiments in Example 2. The divided electrolytic cell was used. Both the anode and the cathode were nickel mesh electrodes. The electrolyte was 0.5M NaOH. The reaction temperature was 20 C. and the current was 22.4 mA. The electrochemical oxidation was continued for 5.6 hours.

(11) The concentration of the materials (number of millimoles per liter) and the results are shown in Table 4.

(12) TABLE-US-00004 TABLE 4 Aldehyde Experiment conversion, No. Feedstock % 9 5 mM benzaldehyde + 45 mM benzoic acid 71.4 10 50 mM 4-CBA + 50 mM terephthalic acid 81.6

(13) The results show that also benzoic acid and terephthalic acid can conveniently be purified by applying the process according to the invention.

EXAMPLE 5

(14) Five more experiments were performed in substantially the same way as described for the experiments in Example 1. The electrolyte was 0.5 M NaOH solution. The feedstock was a solution of 50 mM crude FDCA (containing 1% wt FFCA) per liter of 0.5 M aqueous NaOH. The anode was either a stainless steel plate (Exp. No. 11), a tin plate (Exp. No. 12), a copper mesh (Exp. No. 13), carbon paper (Exp. No. 14) or a platinum mesh (Exp. No. 15). The cathode was a nickel mesh, as used in Exp. No. 2. The reaction temperature was 20 C. The current applied amounted to 22.4 mA.

(15) The total conversion after 5.6 hours was recorded for Exp. Nos. 11, 14 and 15, whereas the total conversion of aldehyde for Exp. No. 12 was reached after 3.2 hours and for Exp. No. 18 already after 0.4 hours. The results are summarized in Table 5.

(16) TABLE-US-00005 TABLE 5 Residence Aldehyde Exp. No. Anode material time, hr conversion, % 11 Stainless steel 5.6 92.3 12 Tin plate 3.2 100 13 Copper mesh 0.4 100 14 Carbon paper 5.6 78.0 15 Platinum mesh 5.6 33.0

(17) The results show that all electrodes enable the oxidation of the aldehyde, although non-noble metals that are used as anode material perform better than noble metals, even when the noble metal is present as anode with a larger surface area. The copper electrode is particularly effective. Also the carbon electrode is significantly more efficient than the platinum electrode.

COMPARATIVE EXPERIMENT

(18) An acid product was obtained from the oxidation of 5-methoxymethylfurfural in acetic acid in the presence of a catalyst that contained cobalt, manganese and bromine. The acid product has precipitated and the solid product was filtered to remove acetic acid. Subsequently, the acid composition washed with water and the amount FFCA therein was determined. The product was subsequently taken up in water at 90 C. and completely dissolved. The weight ratio of acid product to water was about 1:150. The solution was allowed to cool to 20 C., and a precipitate was formed. The precipitate was filtered off and dried. This precipitate was recrystallized two more times, using the above procedure. The yield of solids obtained, based on the weight of the acid product, was determined. The amount of FFCA in the final precipitate was also determined. The results are shown in Table C1, below.

(19) The recrystallization experiment was repeated with the same acid product as starting material, but the product was dissolved in acetic acid at 100 C. The weight ratio of acid product to acetic acid was about 1:150. The solution was allowed to cool to 5 C., and a precipitate was formed. The precipitate was filtered off and dried. The yield of solids obtained after three recrystallizations, based on the weight of the acid product, was determined. The amount of FFCA in the final precipitate was also determined. The results are shown in Table C1, below.

(20) The recrystallization experiment was repeated with the same acid product as starting material, but in the product was dissolved in methanol at 60 C. The weight ratio of acid product to methanol was about 1:26. The solution was allowed to cool to 20 C., and a precipitate was formed. The precipitate was filtered off and dried. The yield of solids obtained after three recrystallizations, based on the weight of the acid product, was determined. The amount of FFCA in the final precipitate was also determined. The results are shown in Table C1 below.

(21) TABLE-US-00006 TABLE C1 FFCA FFCA (acid product), (recrystallized), Solids Yield, Exp. No. Solvent ppmw ppmw % wt C1 Water 6244 3382 53 C2 Acetic acid 6244 1894 60 C3 Methanol 6244 486 15

(22) These results show that recrystallization only has a modest effect on the removal of FFCA from the FDCA product whilst the loss of FDCA product is considerable.